The relationship between multi-joint eccentric strength and change of direction performance : A cross-sectional study on female soccer players

(1)

The relationship between multi-joint

eccentric strength and change of

direction performance

A cross-sectional study on female soccer players

Michell Dahlin

THE SWEDISH SCHOOL OF SPORT AND HELATH SCIENCES Master’s degree Project 45:2020 Master of Sports Science: 2017-2020 Supervisor: Toni Arndt Examiner: Magnus Lindwall

(2)

Acknowledgements

I would like to thank my supervisor, Professor Toni Arndt. Your innovative ideas and feedback made this thesis into an amazing project that has broaden my perspectives on biomechanics.

A special thanks to Olga Tarassova from the Biomechanics and Motor Control laboratory. This study wouldn’t have been possible without your help!

Thank you Ondrej Spiegl for critically reviewing my thesis during its final stages.

To all the participant who voluntarily chose to take part in this study, thank you!

Finally, I would like to thank Rebecka. Your everyday love, encouragement and support is unparalleled.

(3)

Abstract

Aim: The purpose of this study was to investigate the relationship between a multi-joint

eccentric strength test with both change of direction performance and ground reaction force (GRF) of the penultimate foot contact (PFC) in female soccer players. This study also aimed to investigate the bilateral strength and performance differences between the dominant and non-dominant limb.

Method: 21 female soccer players volunteered to participate in this study. All participants

were free from injury and were currently playing in the 3rd_{highest league in Sweden. All tests}

for each participant took place during one visit at the Swedish School of Sport and Health Sciences. Strength measurements were performed on an isokinetic dynamometer that was converted into a motor driven leg press machine. Each participant performed three eccentric contractions at three different speeds (400, 800 and 1200 m/s) with each leg in the

dynamometer. To determine change of direction performance, each participant executed three repetitions on each side in a change of direction test. During the change of direction test, force plates collected GRF data from the PFC.

Results: Peak propulsive force had a large negative correlation with both eccentric impulse at

400 mm/s (r = - 0,552, p= 0,009) and eccentric impulse at 1200 mm/s (r = - 0,552, p= 0,007) for the right limb. A significant difference in eccentric peak force (p= 0,036) was found between the dominant and non-dominant limb. No significant correlation could be found between eccentric strength at any of the three speed levels and change of direction performance. No correlation could be found between the different GRF components and change of direction performance.

Conclusion: Multi-joint eccentric strength did not significantly correlate with change of

direction performance. Moreover, eccentric strength did not correlate with the GRF during a change of direction. Collectively, these findings are not in line with previous investigations.

(4)

Sammanfattning

Syfte och frågeställningar: Syftet med denna tvärsnittsstudie var att undersöka sambandet

mellan excentrisk styrka och riktningsförändringsförmågan samt normlakrafterna vid en riktningsförändring hos kvinnliga fotbollsspelare. Vidare undersökte denna studie även bilaterala skillnader styrka och prestationsförmågan i ett riktningsförändringstest mellan det dominanta respektive icke-dominanta benet.

Metod: 21 kvinnliga fotbollsspelare rekryterades som frivilliga deltagare i studien. Samtliga

forskningspersoner var skadefria och spelade vid tidpunkten för studien aktivt i Sveriges 3:e högsta liga för damer. Alla tester utfördes under ett besök på Gymnastik- och

idrottshögskolan i Stockholm. En isokinetisk dynamometer modifierades till en motordriven benpress maskin. Varje försöksperson genomförde tre stycken excentriska kontraktioner vid tre olika hastigheter (400, 800 och 1200 m/s) med respektive ben. Vidare genomförde varje försöksperson tre repetitioner i ett riktningsförändringstest med respektive ben för att kvantifiera riktningsförändringsförmågan. En kraftplatta registrerade reaktionskraften från näst sista fotisättningen under riktningsförändringstestet.

Resultat: Peak propulsion force hade ett starkt negativt samband på höger ben med både

eccentric impuls på 400 mm/s (r = - 0,552, p= 0,009) och på 1200 mm/s (r = - 0,552, p= 0,007). Det fanns en signifikant skillnad mellan det dominanta och icke-dominanta ben för eccentric peak force (p= 0,036). Ingen signifikant korrelation existerade mellan excentrisk styrka på några av de tre hastigheterna och riktningsförändringsförmågan. Vidare kunde ingen signifikant korrelation hittas mellan normlakrafterna vid en riktningsförändring och

prestationsförmågan vid ett riktningsförändringstest.

Slutsats: Excentrisk styrka hade inget signifikant samband med prestationsförmågan i ett

riktningsförändringstest. Excentrisk styrka hade inte heller något signifikant samband med reaktionskrafterna vid en riktningsförändring. Sammanfattningsvis är dessa resultat inte i linje med tidigare studier.

(5)

1. Introduction

Soccer is an intermittent sport in which players perform various high intensity actions interspersed by varying rest periods that are characterized by lower intensity activities (Bradley & Noakes, 2013; Pareja-Blanco et al., 2016; Carling et al., 2012). The amount of high intensity actions such as sprints, changes of direction, jumps as well as distance covered in high intensity zones have increased through the years (Barnes et al., 2014). While

performing high speed running is an integral part of soccer, so is the ability to effectively change direction. Studies have reported that soccer players perform nearly 700 directional changes during one match. Additionally, change of direction performance is recognized as a crucial factor to compete at the highest level in soccer (Bloomfield et al., 2007; Fessi et al., 2018; Mirkov et al., 2008; Schimpchen et al., 2016).

Change of direction and agility has previously been synonymous to each other, however, Sheppard & Young (2006) suggest that these two are to be considered as two separate skills. This is based on the cognitive components of decision making and visual scanning that contribute to agility performance in sport (Young et al., 2002). Agility is therefore defined as

“a rapid whole-body movement with a change of velocity or direction in response to a stimulus” (Sheppard & Young, 2006, p. 923). On the contrary, change of direction does not include a perceptual and decision-making component and is therefore defined as the ability to decelerate, change movement direction and accelerate again (Sheppard & Young, 2006). This study will hereinafter use the definitions for change of direction and agility as suggested by Sheppard & Young (2006).

Sheppard & Young (2006) suggest that technique, straight sprinting speed and leg muscle qualities are key factors for change of direction performance. Moreover, leg muscle qualities include strength, power and reactive strength (Figure 1). This notion is supported by

investigations that found correlations between straight sprinting, technique, leg muscle qualities and change of direction performance (Jones, Bampouras, & Marrin, 2009; Spiteri et al. 2013; Spiteri et al., 2014). Furthermore, a variety of leg muscle actions are necessary for a successful change of direction as eccentric strength (braking phase), isometric strength (plant phase) and concentric strength (propulsive phase) will allow the subject to decelerate and reaccelerate into a new direction more efficiently (Spiteri et al., 2014).

(8)

Figure 1 – Components of a change of direction as suggested by Sheppard & Young (2006).

1.1 Eccentric strength

When the magnitude of force that is applied to a muscle exceeds that produced by the muscle itself, the muscle will lengthen. The active lengthening of the muscle is referred to as an eccentric muscle contraction (Lindstedt et al., 2001). The force that can be exerted by the muscle when lengthening will be at its greatest when the speed is increased. Contrary, the exerted force will be at its lowest when the speed is decreased (Figure 2).

Figure 2. Force-Velocity relationship. Redrawn from McGinnis, (2013, p. 294).

Change of direction

Leg muscle qualities

Technique Straight sprinting speed

(9)

The stretch- shortening cycle function appears to be particularly responsive to eccentric stimuli as suggested by Douglas et al. (2017). The stretch- shortening cycle is an active stretch of a muscle followed by an immediate shortening of the same muscle. It is well known that the use of the stretch- shortening cycle will result in greater force production than

concentric only movements (Bobbert et al., 1996). The relationship between power and the stretch- shortening cycle could be explained by stored elastic energy within the

musculotendinous unit, which occurs during the lengthening phase of the muscle. When the muscle begins to concentrically contract after the eccentric action, the stored energy will be released, allowing for a greater force- and power production (van Ingen Schenau et al., 1997).

Eccentric strength training has been utilized for both injury prevention training and strength performance (Askling et al., 2003; Roig et al., 2009). For instance, Askling et al., (2003) found a significant reduction in hamstring injury occurrence among professional soccer players compared to a control group after 10 weeks of strength training with an emphasis on eccentric work. Roig et al., (2009) showed that resistance training with an emphasis on the eccentric phase of the lift, had superior neuromuscular adaptions compared to when the emphasis was on the concentric phase of the lift. Moreover, Papadopoulos, et al. (2014) found that 8-weeks of high load eccentric strength training increased maximal concentric and

eccentric force, maximal power and drop jump performance.

1.2 Eccentric strength and change of direction

As the angle of a change of direction increases from 45° to 90°, the approach velocity before turning will decrease, suggesting that sharper directional changes require higher amount of braking forces (Hader, Palazzi, & Buchheit 2015). For instance, soccer players had higher peak deceleration and longer distance at peak deceleration when comparing a 90° cut to a 45° cut. Furthermore, peak acceleration and distance at peak acceleration were lower during the 90° cut. The lower approach velocities observed when cutting angle increases are most likely due to greater braking forces necessary to decelerate more effectively.

Jones et al. (2017) investigated how bilateral eccentric knee extensor and flexor torque

correlated to a 180° change of direction in female soccer players. The results showed that both eccentric knee extensor (r= -0,674, p< 0,01) and flexor (r= -0,603, p< 0,05) strength had a high correlation with change of direction performance. Furthermore, a moderate correlation was observed in Jones et al., (2017) between eccentric knee extensor strength and the change

(10)

in velocity from the start of the penultimate foot contact (PFC) to the final foot contact (FFC) (r= -0,562), highlighting the value of eccentric strength for decelerations. Spiteri et al., (2014) found a similar relationship between bilateral eccentric strength and change of direction performance in female basketball players (r= 0,942, p= 0,001).

To the author’s knowledge, only one study has examined the relationship between unilateral eccentric strength and change of direction performance. Thomas et al. (2018) studied the relationship between eccentric unilateral knee extensor torque and change of direction performance. A low-to-moderate correlation was found between change of direction performance and eccentric strength (r= -0,30 to -0,45, p ≤ 0.05). The same authors also investigated how a change of direction deficit correlated with the same variables. Change of direction deficit refers to the additional time that one directional change requires when compared with a pure linear sprint over an equivalent distance (Nimphius et al., 2016). However, no significant correlation was found (r= -0,05 to 0,20). A possible explanation for the results found by Thomas et al., (2018) might be that only one muscle group was tested (knee extensor). Although eccentric knee extensor strength has been shown to correlate with change of direction performance (Jones et al. 2017), kinematic assessments indicate that a coordination between hip and knee musculature is required for an effective directional change (Green et al., 2011; Spiteri et al., 2013).

As an effective directional change is composed of rotations about several joints (e.g. hip, knee and ankle) working together simultaneously (Havens & Sigward, 2015a), eccentric strength for change of direction performance should be examined as such (i.e. multi-joint). Spiteri et al., (2014) found a high correlation between an eccentric free-weight back squat and change of direction performance. A limitation in Spiteri et al., (2014) might be that the resistance was isotonic (constant external resistance) and not isokinetic (constant speed). Using isokinetic testing would provide the subjects with the possibility to exhibit maximal force, independent of joint position. Furthermore, as directional changes occur during high speeds (Spiteri et al., 2013), there is a need to examine multi-joint eccentric strength during high speeds to gain a better understanding of its importance for change of direction performance.

(11)

1.3 Injuries associated with change of direction

The majority of injuries in soccer occur during non-contact situations, meaning that there is no physical contact with another player at the time of the injury (Kaneko et al., 2017). This is true for both male and female athletes (Boden et al., 2000; Faunø & Wulff Jakobsen, 2006). A systematic review from Shimokochi & Shultz (2008) showed that the most common

mechanism of an anterior cruciate ligament (ACL) injury in female athletes occurred during a deceleration with the knee somewhat flexed combined with a dynamic valgus rotation.

Moreover, the body weight was transferred over to the final foot contact (FFC). Studies have found women to have a six times higher risk of an ACL injury then men (Arendt & Dick, 1995; Stanley et al., 2016). An ACL injury does not only result in short- and long-term physical weakening, but also causes personal and professional impairment for the athlete (Gottlob & Baker, 2000; Yu & Garrett, 2007).

Dos’Santos et al. (2018) suggest that in order to execute faster changes of direction, knee joint loading will have to increase, making it a “performance-injury-conflict”. This is supported by Schreurs et al., (2017) who that found that as the changes of direction got faster and sharper, greater knee valgus angle and reduced knee flexion was observed.

Although the FFC leg is the most common limb to injure during a change of direction (Shimokochi & Shultz, 2008), force characteristics of the PFC may play a vital role in reducing knee joint loading in the FFC. For instance, Jones et al., (2016) found that when the horizontal braking forces were higher in the PFC than in the FFC, knee abduction moment was reduced in female soccer players. Hence, if subjects are able to achieve greater braking forces in the PFC, it may reduce the momentum and decrease the joint loading during the FFC.

1.4 Bilateral differences

As previously stated, change of direction performance is determined by several leg muscle qualities such as strength, power and reactive strength (Sheppard & Young, 2006). Due to the unilateral nature of running, a deficiency in the ability to generate force in one leg could limit an athlete´s ability to accelerate, decelerate and change direction (Meylan et al., 2009). Lockie et al., (2012) investigated the relationship between bilateral strength differences of the knee flexor and extensor musculature with change of direction. A significant correlation was found between eccentric knee flexion strength differences and change of direction performance and this was true for both torque (r= 0,669, p= 0,005) and work (r= 0,638, p= 0,008). However, no

(12)

significant correlation could be found for bilateral eccentric knee extension differences and change of direction performance. The authors of the study discuss whether or not an interlimb strength difference plays a vital role for change of direction performance or not. It is

hypothesised that in order to overcome such an imbalance, some athletes may be able to compensate for the weaker limb by favouring the stronger limb.

Although a strength difference between limbs may not be crucial for performance, studies have shown that executing a change of direction on the non-dominant limb may lead to disadvantageous outcomes. For instance, Hart et al. (2014) investigated how change of direction performance was affected when executed on the dominant leg versus the non-dominant leg during multidirectional tests. The results showed that when subjects performed the change of direction with their dominant leg, the completion time of the task was

significantly decreased (8,0-8,2%; 0,72-0,73 s ; p= 0,001), indicating that the selected leg used to perform the change of direction will impact the time needed to complete the task. Similar time performance differences when utilizing the dominant leg were also evident for 45°, 90°, 135° and 180° changes of direction in Rouissi et al., (2016). These results imply that certain unilateral leg muscle qualities may dictate performance outcomes.

1.5 The penultimate foot contact and change of direction

The force that acts on a body as a result of it hitting the ground is referred to as the ground reaction force (GRF) (McGinnis, 2013, p. 21). When discussing change of direction, the penultimate foot contact (PFC) and final foot contact (FFC) are of most interest. Several key characteristics during a change of direction have been attributed to the PFC and FFC

respectively.

Many studies have focused on the FFC (Dempsey et al., 2007, 2009; Marshall et al., 2014; Spiteri et al., 2013, 2015) and its association with both injury prevention and performance. However, a growing number of investigations have now started to examine the PFC. This is based on the idea that directional changes occur during multiple steps, where the decelerations are initiated several steps prior to the FFC in order to reduce momentum (Andrews et al., 1977). Moreover, both Mornieux et al., (2014) and Wheeler & Sayers, (2010) found, based on kinematic data, that subjects make small anticipatory postural adjustments in the step prior to the FFC.

(13)

Havens & Sigward, (2015b) compared the different GRF parameters between the PFC and the FFC during a 45° and 90° change of direction and found a significant higher posterior braking force and posterior impulse (p= 0,001) for the PFC. Similar results have been described in 180° turns. Graham-Smith et al., (2009) found that both the peak horizontal (p= 0,02) and vertical (p=0,01) braking forces were higher in the PFC than in the FFC. Jones et al., (2016) found horizontal impulse (p=0,024), peak horizontal braking force (p=0,045) and peak vertical force (p <0,001) to be significantly higher in the PFC compared to the FFC.

Spiteri et al., (2015) found that faster subjects had shorter ground contact times during a change of direction test, additionally, stronger subjects also spent less time braking then their slower counterparts. Therefore, shorter ground contact times indicate that the subjects spend less time braking and instead propel themselves into the new direction. The propulsive characteristics of a change of direction has only been studied in the FFC (Spiteri et al., 2013) and its role in the PFC is not yet understood. As the angle of the cut increases, so will the ground contact time as suggested by Havens & Sigward, (2015b). However, the

abovementioned studies only examined the FFC. DosʼSantos et al., (2017) compared the ground contact time of the PFC between fast and slower subjects, but were unable to find any significant difference. Havens & Sigward (2015b) implied that as the angle of the cut

increases, so will the ground contact time as a greater braking force is needed.

To the authors knowledge, only two studies have analyzed the association between change of direction performance and force characteristics of the PFC. Graham-Smith et al., (2009) found a strong correlation between horizontal braking force of the PFC and faster change of direction (r= 0,647, p= 0,016). DosʼSantos et al., (2017) found a moderate correlation between horizontal braking force and faster changes of direction. However, this was not true when the directional change was executed on the other limb (r=0,046).

(14)

1.6 Description of the problem area

As indicated above, eccentric strength is an important part of change of direction

performance. The studies that have investigated the relationship between change of direction and eccentric strength have utilized isokinetic testing in dynamometers and have been limited to the FFC. However, none of the abovementioned studies have investigated the relationship between the eccentric strength of a multi-joint exercise with change of direction performance. Given that a change of direction task is composed of several muscle groups and joints

working together simultaneously (Green et al., 2011; Spiteri et al., 2013), a multi-joint isokinetic test could be considered the most suitable method for a valid assessment. The need of multi-joint isokinetic tests has been previously suggested (Jones et al., 2017; Spiteri et al., 2014). There is therefore a need to investigate the role of eccentric strength with change of direction performance.

1.7 Purpose

The purpose of this study was to investigate the relationship between multi-joint eccentric strength at different speeds with both change of direction performance and GRF of the penultimate foot contact in female soccer players. This study also investigated the bilateral strength and performance differences between the dominant and non-dominant limb. Knowledge of this would assist strength and conditioning coaches in designing more appropriate training programs.

This study tried to answer the following questions:

1. Is there a relationship between eccentric force and change of direction performance on the given limb?

2. Is there a relationship between eccentric force and GRF on the given limb? 3. Is there a relationship between GRF and change of direction performance on the

given limb?

4. Does eccentric strength, change of direction performance time and GRF vary between the dominant limb and non-dominant limb?

(15)

2. Method

2.1 Study design

This investigation emanated from a quantitative approach and had a cross-sectional study design. During the study, nothing was done to alter or change the behavior of the participants. The participants completed the test-protocol during one two-hour visit at the Swedish School of Sport and Health Sciences (GIH). Each participant was instructed to not perform any resistance training or high intensity work during the last 24 hours before partaking in the study.

2.2 Subjects

21 female soccer players participated in this study (19,1 ± 2,0 years; 63,9 ± 6,5 kg; 170,3 ± 4,4 cm). Inclusion criteria were to be actively playing at a level of at least Division 1 (3rd

highest in Sweden) and a minimum age of 16. Furthermore, all players needed to be free from lower limb injuries for the past six months and be training under normal conditions with their respective teams. A number of female soccer teams within the greater area of Stockholm were contacted, either by telephone or by email, and informed about the study. All participants were informed of the purpose of the study, testing protocol and potential risk factors and given the opportunity to ask any questions concerning the study.

2.3 Procedure

Each participant attended the laboratory of Biomechanics and Motor Control (BMC) on one occasion. The participants were instructed to come well rested, wearing shorts that didn’t cross over the knees, and indoor running shoes. Upon arrival, each participant was again informed about the research purposes and gave their written consent to participate in this study. Once the consent form was signed, the test procedure could officially begin. The first step included a warmup consisting of 10 minutes of cycling (100W) on an ergometer (Monark 829, Varberg, Sweden).

2.3.1 Eccentric strength test

An isokinetic dynamometer (IsoMed 2000-system, D&R Ferstl GmbH, Hemau, Germany) with an additional linear adapter that converts it into a motor driven leg press was utilized for the multi-joint eccentric strength assessment (Figure 3). The standardization followed the guidelines of Dirnberger et al. (2013). The participants were seated in the dynamometer chair in an upright position and with a hip joint angle of 75°. The footrest was positioned 10 cm

(16)

above the dynamometer chair and the heel was placed on its lower marginal frame. The footrest was inclined in 15° plantar flexion. Seat belts were applied across the shoulders and thorax to minimize upper-body movements. The range of motion (ROM) was set between 25– 90° of knee flexion using a handheld mechanical goniometer. The trochanter major, the lateral femoral epicondyle and the lateral malleolus served as reference points and were found by palpation and then marked. Once the correct ROM was established, a portable hard knee rest was positioned under the participant’s knee joint to prevent overextension of the knee. Before using the dynamometer, instructions were given to the participant regarding safety issues and the procedure.

The leg-press protocol in the dynamometer consisted of three maximal eccentric contractions for each leg at three different speeds (400, 800 and 1200 mm/s). Prior to testing the

participants were allowed as many repetitions needed to be accustomed with the

dynamometer. The participant performed three repetitions at one speed with one leg, then changed leg and performed three repetitions at the same speed, before changing speed. The given speed and order of legs (left/right) was randomly chosen beforehand using Excel´s RANDBETWEEN function. Between every repetition, each participant was allowed 60 s of passive rest. To start the dynamometer, the participant had to push concentrically against the footrest with a force of 400 N. This starting procedure was established both as a safety

precaution and to eliminate reaction time characteristics. The dynamometer force and position data were sampled at 3000 Hz. The analog signal was converted into a digital signal using a CED power 1401 data acquisition system (version 7.0, Cambridge Electronic Design, UK) and collected into Spike 2 software (version 7.09a, Cambridge Electronic Design, UK). The peak value of the three trials was used for further analysis. Limb dominance was defined as the limb that generated the greatest amount of eccentric peak force in the dynamometer during all three speeds. The variables included within the eccentric strength test analysis were peak force (N·kg−1_{) and impulse (N·s}−1_·kg−1_).

(17)

Figure 3 – Set-up of the converted isokinetic dynamometer.

2.3.2 Change of direction test

All change of direction tests took place in the BMC laboratory fitted with a 1.22 m wide tartan running surface. For the change of direction test, participants were instructed to sprint to a line marked four meters from the starting line, placing the left or right foot on the line (depending on the trial), turn 180° and sprint back four meters to where they started (Figure 4). A left- or right sided change of direction was defined as the following;

Right change of direction = left FFC and right PFC Left change of direction = right FFC and left PFC

Photocells (Brower TC Timing System, Draper, Utah, USA) were placed at the start- and finish line, at a height of 0,8 m and separated by 3 m as recommended by Haugen & Buchheit (2016). To avoid a premature triggering of the initial photocell, each participant stood 0,5 m behind the first photocell (Thomas et al., 2018). Each start was executed from a standing

(18)

position with the feet parallel and arms crossed over the shoulders. Two force plates (Kistler 9281EA, AG Winterthur, Switzerland) were embedded into the running surface to measure GRF of the penultimate foot contact with a sampling frequency of 2500 Hz. GRF data were collected in Qualisys Track Manager (QTM, Qualisys AB, v.2.14).

If the participants changed direction before stepping past the 4-m mark, turned off the incorrect foot or didn’t entirely step on any of the force plates, the trial was disregarded and repeated after a recovery period. Prior to testing, the participants were allowed as many repetitions needed to be accustomed with the change of direction and foot placement. Each participant performed three trials on each side that had been randomly chosen with Excel´s RANDBETWEEN function, separated by one minute rest. The peak value of all three trials was used for further analysis. The average time of all three trials was calculated for

determining change of direction performance. The variables included for analysis were: - Duration of braking phase (s)

- Duration of propulsion phase (s) - Braking impulse (N·s−1_·kg−1₎

- Horizontal braking impulse (N·s−1_·kg−1₎

- Propulsive impulse (N·s−1_·kg−1₎

- Peak braking force (N·kg−1₎

- Peak horizontal braking force (N·kg−1₎

- Peak propulsive force (N·kg−1₎

(19)

Figure 4 - Change of direction performance test set-up.

2.4 Specification of measuring devices

2.4.1 Brower Timing System

All change of direction sprint times were measured using single-beamed photocells (Brower TC Timing System, Draper, Utah, USA). The single-beamed photocell consists of a

transmitter emitting an infrared beam to a reflector (placed on the opposite side) that reflects the beam back to the transmitter. Inappropriate height adjustment of a photocell may increase timing error, e.g. if mounted at chest height, the beam might be broken twice by the arms and torso separately as suggested by Cronin & Templeton (2008). Therefore, the photocells should be placed at a height at which only one body part may break the beam (Yeadon et al., 1999). Additionally, the distance between photocells should be as large as practically possible (Yeadon et al., 1999).

2.4.2 IsoMed 2000-system

An isokinetic dynamometer (IsoMed 2000- system, D&R Ferstl GmbH, Hemau, Germany) was converted into a motor driven leg press with an additional linear adapter. This allowed the rotary movement of the dynamometer shaft to be converted into a linear movement through an inbuilt tooth belt linear drive. The footrest of the leg press adapter was positioned on the tooth belt. The force applied by the participant against the footrest was detected

(20)

through two bilateral force sensors placed directly at the respective left and right backside of the footrest. Two freely selectable gears on the linear adapter could to be connected to the dynamometer unit through the drive shaft. The two gear ratios (gear I: 1/1; gear II: 1/1.5), determined the maximum linear speed. Gear I had a maximum linear speed of 800 mm/s compared to gear II that could reach 1200 mm/s. For the present study, gear II was used.

2.4.3 Kistler force plate

In order to measure the GRF exerted by the body during a change of direction, two 0,6 m x 0,4 m force plates (Kistler 9281EA, AG Winterthur, Switzerland) were used. Both force plates were firmly attached into the laboratory floor with their upper surfaces parallel to the floor surface. A piece of tartan mat was cut out and attached to the force plate with double sided tape so that it was level with the running surface. A small gap (approximately 5 mm) was free between the tartan on the force plate and the running surface in order to prevent forces being transmitted to the surrounding tartan mat.

The force plate gives an electrical signal proportional to the applied GRF acting upon it. Forces are represented by three dimensional vectors with a vertical (Fz) and two shear

components acting along its surface. The two shear forces act in an anterior-posterior (Fx) and a medial-lateral (Fy) direction.

2.5 Ethics

All participants were informed about the study´s purpose and potential risk factors associated with their participation. Moreover, all participants were given the opportunity to ask further questions regarding the study and had to sign a written informed of consent before the procedure could begin. This study was approved by the Swedish Regional Board of Ethics (Dnr: 2019-01817).

2.6 Validity and reliability

To ensure a high standardization throughout the entire procedure, a test protocol was

systematically followed. In order to establish a proper test protocol, several pilot studies were conducted beforehand. The pilot studies ensured that the measuring devices worked in

accordance with the study´s purpose and that the test leader got accustomed with both the protocol and the measuring devices. Throughout the study, the same test leader was in charge for all the tests, to ensure a high standardization. Furthermore, all participants were given the same instructions during the tests.

(21)

Haugen & Buchheit (2016) recommend that a single-beamed photocell should be avoided if possible, during sprinting, and instead use a dual-beam photocell to avoid arms and legs from triggering the beam. This notion is also supported by Haugen, Tønnessen, Svendsen, & Seiler (2014) who found a 0,03 second standard error of measurement (SEM) and a ~2% CV when investigating the single-beams reliability. Additionally, an absolute difference of -0,05 – 0,06 s exist between single- and dual beam timing systems (Haugen et al., 2014). To avoid that an early or late triggering would solely determine the results during a single-beamed set up, the average time of three trials was calculated.

The IsoMed 2000-system reliability has been investigated by both Dirnberger, Kösters, & Müller (2012) and Dirnberger, Wiesinger, & Müller (2012). Although both of these studies found the IsoMed 2000-system’s reliability to be high, these studies were limited to single-joint rotational movements. Only one study (Dirnberger et al., 2013) has evaluated the IsoMed 2000-system using the multi-joint (leg press) movements with a linear adapter. The authors found a moderate to strong reproducibility (intraclass correlation coefficient (ICC): 0,823-0,951) for the leg press exercise in a test-retest comparison of peak force during three identical test sessions. However, a significantly (p= 0,05) higher peak force value was observed by the participants between test day one and two, indicating that a practice-based improvement may occur (Dirnberger et al., 2013). Due to time limitations in this study, no separate familiarization session was performed. To counteract this issue, all participants were allowed more warm-up repetitions (≈20) compared to the participants in (Dirnberger et al., 2013) who only got four warm-up repetitions.

Force plates are generally referred to as the “gold standard” for the direct acquisition of GRF data (Nigg & Herzog, 2006, p. 324). Studies reporting upon GRF during changes of direction almost exclusively use force plates (Dos’Santos et al., 2018).

2.7 Data- and statistical analysis

Kinetic data from the force plates were analyzed in Visual3D (C-Motion Inc, Germantown, USA, v.7), whereas kinetic data from the IsoMed 2000-system were analyzed in Matlab (R2015b, Mathworks Inc, USA). Custom design scripts were written in both programs. The braking phase of the change of direction was defined as the first instance at which 10 N was reached with a peak of at least 300 N, until the instance when the force was below 10 N. The propulsion phase was defined similarly, except that is was calculated from the second

(22)

instance at which 10 N was reached. If the force did not decrease below 10 N, the intersection between braking and propulsion phase was defined as the point of minimum force (Figure 5).

Figure 5- Illustration of the braking- and propulsive phase in a force-time curve during a change of direction.

Peak force and impulse were defined as the resultant of the vertical, medio-lateral and anterior-posterior GRF (√Fx2_+Fy2_+Fz2_{), whereas horizontal peak force and impulse was}

defined as the resultant of the medio-lateral and anterior-posterior GRF (√Fx2_+Fy2_{) as}

suggested elsewhere (DosʼSantos et al., 2017) (Figure 6).

Figure 6 – Example of the different GRF on a force-time curve during a change of direction. A) Anterior-posterior (Fx), medio-lateral (Fy) and vertical (Fz) components of the GRF. B). The resultant force vector (√Fx2_+Fy2_+Fz2_{) of the} GRF.

Sprint time data from the Brower timing system were first manually registered and then exported to Excel for further analysis along with kinetic data from Visual3D and Matlab. In Excel, tables were created for every participant and all kinetic variables were normalized to body weight. Next, peak values for force, impulse, GRF and average sprint time were calculated to the nearest 0.001 s and selected for further analysis.

A

(23)

Statistical analysis was performed in the IBM SPSS (Statistical Package for the Social Sciences, v22) statistical package for Windows (SPSS, Chicago, IL, USA). Descriptive statistics (mean ± SD) were presented for each variable. A Shapiro-Wilk test was used to check if the data were normally distributed. To examine the eccentric strength differences between legs, a paired sample t-test was conducted if the data were normally distributed, and a Wilcoxon Signed Rank test if non-parametric. To examine the relationship between

eccentric strength variables and time during the change of direction, Pearson’s product moment correlation coefficient was utilized if the data were normally distributed and Spearman’s rank order correlation coefficient if the data were non-parametric. Correlations were evaluated as follows: small (0,10 – 0,29), moderate (0,30 – 0,49), large (0,50 – 0,69), very large (0,70 – 0,89), nearly perfect (0,90 – 0,99), and perfect (1,0) (Hopkins, 2000). Statistical significance was set at p ≤ 0.05.

Power analysis was performed a priori to determine the sample size required to produce an effect size of 0,9, a power of 0,8 at an alpha level of 0,05 using a statistical power analysis software (G*Power v.3.1.9.3, Germany). An effect size of 0,9 is considered as large by Cohen´s effect size score. To achieve statistical power, a sample size of 21 participants was required.

(24)

3. Results

3.1 Relationship between eccentric strength and change of

direction

Pearson’s correlation was used to examine the relationship between eccentric strength

parameters and change of direction performance. However, no significant correlation could be found for any of the variables (Table 1).

Table 1 - Relationship between eccentric strength parameters and change of direction performance.

3.2 Relationship between GRF and change of direction

To determine the correlation between the different GRF parameters of the PFC and change of direction performance, Pearson’s correlation was used. No significant correlation could be found for any variable (Table 2).

Table 2 - Relationship between GRF and change of direction performance.

r p r p Peak force 400 mm/s -0,144 0,546 -0,039 0,870 Peak force 800 mm/s -0,209 0,376 -0,248 0,279 Peak force 1200 mm/s -0,197 0,393 -0,094 0,695 Impulse 400 mm/s -0,212 0,357 -0,078 0,738 Impulse 800 mm/s -0,140 0,545 -0,210 0,362 Impulse 1200 mm/s -0,274 0,229 -0,127 0,585

COD Left COD Right

r p r p

Duration of braking phase (s) -0,084 0,724 0,255 0,264

Duration of propulsion phase (s) 0,062 0,790 0,259 0,258

Braking impulse (N·s−1_·kg−1₎ _-0,137 _0,553 _0,090 _0,698

Horizontal braking impulse (N·s−1·kg−1) -0,361 0,108 0,079 0,742

Propulsive impulse (N·s−1_·kg−1₎ _0,118 _0,611 _0,242 _0,291

Peak braking force (N·kg−1₎ _-0,042 _0,858 _0,059 _0,799

Peak horizontal braking force (N·kg−1₎ _0,059 _0,798 _-0,185 _0,421

Peak propulsive force (N·kg−1₎ _0,409 _0,066 _0,116 _0,618

Peak vertical braking force (N·kg−1₎ _-0,036 _0,877 _0,080 _0,730

COD Left COD Right

COD = change of direction.

(25)

3.3 Relationship between eccentric strength and GRF

Using Pearson’s correlation, peak propulsion force of the right PFC had a large negative correlation with both eccentric impulse at both 400 mm/s (r = - 0,552, p= 0,009) and 1200 mm/s (r = - 0,552, p= 0,007). No other significant correlation could be found between any other variables (Table 3 & 4).

(26)

Table 3- Relationship between eccentric strength and GRF on the left PFC. r p r p r p r p r p r p Du ra tio n o f b ra kin g pha se (s ) 0,002 0,994 -0 ,1 00 0,684 -0 ,1 03 0,665 -0 ,0 84 0,725 -0 ,3 76 0,102 0,042 0,860 Du ra tio n o f pr opul sion pha se (s ) -0 ,1 11 0,640 -0 ,2 19 0,353 -0 ,0 16 0,944 0,129 0,578 -0 ,1 01 0,662 -0 ,0 11 0,961 Br ak in g i m pu lse (N ·s −1 ·k g −1 ) 0,118 0,619 -0 ,0 71 0,768 -0 ,1 66 0,471 0,075 0,748 0,345 0,126 0,056 0,809 Ho riz on ta l b ra kin g im pu ls e ( N ·s₋₁ ·k g −1 ) 0,160 0,501 0,086 0,718 0,038 0,870 0,128 0,581 0,317 0,162 0,117 0,612 Pr op uls iv e i m pu lse (N ·s −1 ·k g −1 ) -0 ,0 59 0,804 -0 ,0 19 0,935 -0 ,1 65 0,475 -0 ,1 84 0,425 -0 ,2 10 0,360 -0 ,1 16 0,617 Pe ak b ra kin g f or ce (N ·k g −1 ) -0 ,1 68 0,479 -0 ,2 61 0,266 -0 ,2 86 0,209 -0 ,0 98 0,673 0,269 0,238 -0 ,0 89 0,700 Pe ak h or iz on ta l br aki ng for ce (N ·k g −1 ) -0 ,1 57 0,508 -0 ,2 23 0,345 -0 ,2 84 0,213 -0 ,0 24 0,917 0,207 0,367 -0 ,1 67 0,469 Pe ak p ro pu lsi ve fo rc e (N ·k g −1 ) -0 ,0 13 0,958 0,095 0,690 -0 ,1 01 0,665 -0 ,3 28 0,147 -0 ,0 73 0,752 -0 ,2 05 0,373 Pe ak v er tic al br ak in g fo rc e (N ·k g −1 ) -0 ,1 72 0,470 -0 ,2 96 0,204 -0 ,2 82 0,215 -0 ,1 01 0,662 0,265 0,246 -0 ,0 82 0,724 Im pu lse 1200 m m /s P eak for ce 400 m m /s P eak for ce 800 m m /s P eak for ce 1200 m m /s Im pu lse 400 m m /s Im pu lse 800 m m /s

(27)

Table 4 - Relationship between eccentric strength and GRF on the right PFC. r p r p r p r p r p r p Du ra tio n o f b ra kin g pha se (s ) -0 ,1 62 0,496 -0 ,3 41 0,130 -0 ,0 22 0,926 -0 ,2 53 0,269 -0 ,1 45 0,531 -0 ,0 62 0,790 Du ra tio n o f pr opul sion pha se (s ) 0,162 0,496 0,214 0,351 -0 ,0 71 0,765 0,370 0,098 0,186 0,421 0,317 0,162 Br ak in g i m pu lse (N ·s −1 ·k g −1 ) -0 ,2 58 0,271 -0 ,1 72 0,456 -0 ,0 44 0,853 -0 ,1 99 0,387 -0 ,0 25 0,915 -0 ,0 18 0,940 Ho riz on ta l b ra kin g im pu ls e ( N ·s₋₁ ·k g −1 ) 0,033 0,894 0,102 0,669 0,139 0,571 0,007 0,975 0,131 0,583 0,158 0,506 Pr op uls iv e i m pu lse (N ·s −1 ·k g −1 ) 0,036 0,881 -0 ,0 98 0,673 -0 ,1 15 0,628 -0 ,1 14 0,621 -0 ,0 78 0,738 -0 ,2 27 0,322 Pe ak b ra kin g f or ce (N ·k g −1 ) -0 ,2 11 0,372 0,037 0,874 0,039 0,869 -0 ,1 16 0,617 0,018 0,939 -0 ,0 44 0,849 Pe ak h or iz on ta l br aki ng for ce (N ·k g −1 ) -0 ,1 54 0,517 0,116 0,615 0,138 0,561 0,044 0,851 0,092 0,693 0,159 0,491 Pe ak p ro pu lsi ve fo rc e (N ·k g −1 ) -0 ,2 50 0,288 -0 ,3 36 0,137 -0 ,2 94 0,208 -, 55 2* 0,009 -0 ,3 75 0,094 -0 ,5 69 * 0,007 Pe ak v er tic al br ak in g fo rc e (N ·k g −1 ) -0 ,2 65 0,258 -0 ,0 19 0,936 -0 ,0 20 0,933 -0 ,1 67 0,469 -0 ,0 78 0,738 -0 ,1 28 0,581 Im pu lse 1200 m m /s P eak for ce 400 m m /s P eak for ce 800 m m /s P eak for ce 1200 m m /s Im pu lse 400 m m /s Im pu lse 800 m m /s * = significant correlation (p < 0,05).

(28)

3.4 Differences between limbs

A paired samples t-test showed a significant difference between the dominant and non-dominant limb in peak force (p= 0,036). No significant difference could be found for any other variable (table 5).

Table 5 - Comparison of eccentric strength and change of direction parameters between the dominant limb and non-dominant limb.

4 Discussion

The aim of this study was to investigate the relationship between a multi-joint eccentric strength test with both change of direction performance and GRF of the PFC in female soccer players. This study also aimed investigated the bilateral strength and performance differences between limbs. Previous studies had identified eccentric strength to be highly associated with change of direction performance. A limitation in those studies was that the eccentric strength assessments were limited to single-joint movements (e.g. knee extension or knee flexion) and to the final foot contact (FFC). This study is the first to investigate eccentric strength across several joints (i.e. multi-joint movements) in an isokinetic dynamometer. A change of direction is a complex movement where several joints and muscle work together simultaneously (Havens & Sigward, 2015a, 2015b).

The main finding in this study was that eccentric strength had no significant correlation with change of direction performance, this was true for both legs at all three different speed levels. Eccentric strength had no significant correlation with any of the GRF parameters for either leg

Mean Std. Mean Std. p-value

Eccentric strength parameters

Peak force (N·kg−1₎ _24,932 _4,927 _21,848 _3,712 _0,036*

Impulse (N·s−1_·kg−1₎ _16,516 _5,104 _14,865 _4,378 _0,288

Change of direction parameters

Change of direction performance (s) 2,766 0,093 2,715 0,084 0,090

Duration of braking phase (s) 0,405 0,058 0,387 0,065 0,602

Duration of propulsion phase (s) 0,328 0,047 0,328 0,048 0,972

Braking impulse (N·s−1_·kg−1₎ _3,038 _0,288 _3,103 _0,437 _0,759

Horizontal braking impulse (N·s−1_·kg−1₎ _1,491 _0,190 _1,526 _0,221 _0,759

Propulsive impulse (N·s−1_·kg−1₎ _3,230 _0,235 _3,221 _0,187 _1,000

Peak braking force (N·kg−1₎ _25,668 _3,805 _25,842 _5,317 _0,977

Peak horizontal braking force (N·kg−1₎ _12,781 _2,526 _13,178 _2,949 _0,704

Peak propulsive force (N·kg−1₎ _17,357 _1,759 _17,193 _1,365 _0,763

Peak vertical braking force (N·kg−1₎ _22,982 _3,769 _22,716 _4,698 _0,814

Dominant limb Non-dominant limb

(29)

and no significant correlation could be found between change of direction performance and GRF. Lastly, there was no significant difference between limbs when the change of direction was executed on either limb.

4.1 Relationship between eccentric strength and change of

direction

Previous studies have found both knee extensor- and flexor torque to have a significant correlation with change of direction (Jones et al., 2017; Thomas et al., 2018). However, there was a need to examine the effects of an isokinetic multi- joint exercise with change of

direction that would better reflect a sport specific situation in which a coordination of hip- and knee muscles working simultaneously is required. Contrary to previous investigations, this study was unable to find any significant correlation between eccentric strength and change of direction performance.

Spiteri et al., (2014) found that a free-weight back squat had a significant correlation to change of direction performance (r = -0,892, p= 0,001). Although a free weight back squat is a multi-joint exercise, the isotonic nature of the back squat makes it difficult to compare the two. Moreover, the free weight back squat is a bilateral exercise compared to the unilateral exercise performed in this study. This could have had an effect on the results due to different muscle activations during an unilateral and bilateral exercise (Eliassen et al., 2018).

To the authors knowledge, this study is, along with Thomas et al., (2018), the only two studies that have investigated the relationship between unilateral eccentric strength and change of direction performance. This study was unable to find any significant relationship between the two variables. Although Thomas et al., (2018) did find a significant correlation between unilateral eccentric strength and change of direction performance, the correlation was only low-to-moderate (r= -0,30 to -0,45). Moreover, unilateral eccentric strength had no significant correlation with change of direction deficit. As suggested by Sheppard & Young, (2006), leg muscle qualities such as eccentric-, isometric- and concentric strength are all necessary components for an effective change of direction. It could therefore be argued that only investigating one muscle action at the time, independently of the others, may not be sufficient to fully grasp the complexity of the task.

(30)

Contrary to previous investigations, this study utilized the linear velocity (mm/s) of the footrest to quantify speed, whereas Jones et al., (2017) and Thomas et al., (2018) used the angular velocity (°/s) of the knee joint. Dirnberger et al., (2013) suggest that a linear velocity of ≈ 400 mm/s would correspond to an angular velocity of ≈ 80°/s within a ROM of 20-95° knee flexion. Given that Jones et al., (2017) and Thomas et al., (2018) used an angular velocity of 30-60°/s, the velocities used in this study were much higher wich could have affected the results.

4.2 Relationship between GRF and change of direction

In contrast to previous investigations, this study could not find any relationship between GRF parameters of the PFC and change of direction performance. Given that several studies (Andrews et al., 1977; Mornieux et al., 2014; Wheeler & Sayers, 2010) classified the PFC as the most important step in order to reduce momentum during a change of direction, and horizontal braking forces to be much greater in the PFC than in the FFC (Graham-Smith et al., 2009; Havens & Sigward, 2015b), it was hypothesized that GRF of the PFC would have a strong correlation with performance times.

Although Graham-Smith et al., (2009) found horizontal braking force to correlate with change of direction performance, the authors did not specify which component of the horizontal GRF was analyzed, which could have had an impact on the results. However, DosʼSantos et al., (2017) defined horizontal force the same as in this study, i.e. the resultant vector of the medio-lateral and anterior-posterior GRF. It should be noted that even if DosʼSantos et al., (2017) found a significant correlation between horizontal braking force and faster change of direction performance, this correlation was only moderate (r = -0,337). Furthermore, the correlation was only significant when the task was executed on one side and not the other, suggesting that more research is needed.

Shorter ground contact times have been related to faster change of direction performance (Spiteri et al., 2013). This study investigated the relationship between ground contact time of the PFC with change of direction performance, however, no significant correlation could be found for either the duration of braking- or the propulsion phase. DosʼSantos et al., (2017) found no significant difference in ground contact time of the PFC when comparing faster- to slower participants. These results could indicate that shorter ground contact time of the PFC

(31)

aforementioned, it could be argued that for an effective change of direction, the PFC will have to produce greater braking forces relative to the FFC (Jones. et al., 2016). As the ground contact time of the FFC have been shown to differentiate between faster and slower participants (Spiteri et al., 2013), the FFC will be used to push the participant into the new direction with minimum ground contact time.

This study could not find any significant correlation between change of direction performance and eccentric impulse, neither braking or propulsive. With regards to the impulse-momentum relationship (derived from Newton´s second law of motion), which suggests that a change in momentum is equal to the impulse applied to an object, a greater impulse would indicate a faster acceleration or deceleration. In terms of change of direction performance, a large impulse would be desirable. Moreover, as Havens & Sigward, (2015b) have shown, a greater change of direction angle will results in an increased impulse. It is therefore surprising why this study could not find any relationship between the two variables. A potential explanation for the different result might be different methodological approach, as the anterior-posterior and medial-lateral forces were analyzed separately in Havens & Sigward, (2015b).

4.3 Relationship between eccentric strength and GRF

A significant negative correlation was found between eccentric impulse at both 400 mm/s (r= - 0,552, p =0,009) and 1200 mm/s (r= - 0,552, p= 0,007) with peak propulsion force of the right limb during the change of direction test, suggesting that the ability to produce greater eccentric impulse has no benefit on the propulsion force produced during the change of direction test. These findings are in direct contrast to previous investigations where eccentric strength have been linked to faster change of directions (Jones et al., 2017; Spiteri et al., 2014). As the stretch-shortening cycle is highly correlated with eccentric strength (Bridgeman et al., 2018), a greater force production is expected during the concentric phase of the

movement (i.e. propulsive phase) after an active stretch of the muscle (eccentric phase). The active stretch allows stored elastic energy within the musculotendinous unit will be released, resulting in an increased force production (van Ingen Schenau et al., 1997).

Spiteri et al., (2013) found that stronger participants were able to produce greater horizontal and vertical impulse and horizontal peak force during the braking phase of the FFC, compared to weaker participants. It was therefore anticipated that, as participants in this study generated greater eccentric forces in the dynamometer, the GRF would increase. However, this study

(32)

could not find any such relationship. Similar results was found in Jones et al., (2017) where horizontal peak force was greater in stronger participants. The different results presented in the studies by Spiteri et al., (2013) and Jones et al., (2017) compared to this study might be due to the use of single-joint exercises in the dynamometer, compared to a multi-joint exercise used in this study.

4.4 Differences between limbs

Based on Sheppard & Youngs (2006) model of components in a change of direction and the unilateral nature of running, executing a change of direction just as well on both sides would be advantageous in terms of performance (Meylan et al., 2009). The results in this study could not find any significant difference in performance time during a change of direction executed with either limb (p = 0,090), which is in contrast to previous studies. Hart et al., (2014) found an 8,0-8,2% difference between limbs when the change of direction was performed with the dominant limb, which was also the case in Rouissi et al., (2016). Furthermore, greater

bilateral knee flexor imbalances have been associated with a reduced performance (Lockie et al., 2012). However, as this study could not find any difference between limbs in the change of direction test. It is possible that, if a muscle imbalance does exist, the participants may have compensated for the weaker limb by adjusting their technique. A computer simulation study by Yoshioka et al., (2010) found that greater lateral movements of the body occurred during a counter movement jump when the weaker leg was used, suggesting that the load is proportionally distributed between the limbs to better execute the task without compromising performance. Moreover, only a 0,7% difference in jump height was observed between the symmetrical- (0,416 m) and asymmetrical (0,419 m) model suggesting that an asymmetry by itself does not determine performance outcome. However, kinematic data were not included in this study making it impossible to draw any conclusions whether the participants adjusted their technique or not. Nonetheless, based on the model by Sheppard and Young (2006), technique is a fundamental part of change of direction performance.

A significant difference existed between the dominant limb and non-dominant limb for eccentric peak force (p= 0,036). However, no other significant result could be found for any other strength variable or GRF. The dominant limb produced the greatest amount of eccentric peak force in the dynamometer. Yet, braking forces during the change of direction test were

(33)

therefore be questioned whether or not eccentric strength affects the braking forces in a change of direction test.

As bilateral strength differences are common in athletes (Kellis et al., 2001; Newton et al., 2006), it was surprising that this study did not find more differences between limbs. The insignificant strength difference between limbs could be considered favorable, as it would allow the participants to perform the change of direction equally well, regardless of which side the task takes place on.

It is worth mentioning that the definition of limb dominance may differ between studies and be task-dependent. For instance, Rouissi et al., (2016) defined leg dominance by the amount of force produced in a maximal isometric voluntary contraction, while Lockie et al., (2012) defined dominant side by their fastest completion time in a change of direction test.

Additionally, soccer players tend to favor their dominant limb to control and manipulate the ball, and the non-dominant limb to provide stability, indicating the difficulty in defining limb dominance without context. As this study defined limb dominance as the limb that generated the greatest amount of eccentric peak force in the dynamometer, it is possible that a different definition could have produced a different result.

4.5 Limitations

A limitation of this study was that no separate familiarization session was done with the dynamometer. Even though this investigation tried to counteract this by allowing the participants a great number of warm-up repetitions, it is still possible that the number of repetitions were not enough to properly accustom the participants with the dynamometer. This could in turn have had an impact on the results.

This study used a single-beamed photocell to time the change of direction performance test. It has been suggested that a dual-beam photocell is to be preferred to avoid an early triggering by the arms or legs. Therefore, it is possible that an early triggering might have occurred which could have had an impact on the results. Also, this study calculated the average time of all three trials. The results may have been different if the best attempt would have been chosen instead of the average.

(34)

The most frequently used test in change of direction studies is the modified 505 agility test, which consists of a 5-meter sprint, turn 180°, followed by a sprint back to the starting line. Due to limited space in the laboratory, only a 4-m sprint could be completed before the 180° turn. The shorter distance made the comparison with previous studies more difficult. The kinetics and kinematics of the participants may have been different because of this.

The isokinetic legpress exercise only measured forces in the sagittal plane. As a 180° change of direction occurs in all three planes of motion, it is possible that some of the eccentric force during the change of direction was produced by muscles acting in a different plane of motion, e.g. abduction and adduction.

To collect GRF data, the participants needed to land on the force plate with their entire foot. This laboratory setting meant that the participants had to aim at the force plate. It must therefore be taken into account that soccer players do not usually aim at where to plant their foot during a change of direction in a match situation. Additionally, the tartan running surface and indoor running shoes may also have influenced the results also differed compared to the actual conditions in soccer.

4.6 Future research

As mentioned earlier, it is possible that the participants adjusted their technique in order to perform the task with equal efficacy when executing the change of direction with different limbs. Future research should investigate the different kinematics of the PFC together with muscle activation of different muscle groups. From an injury prevention perspective, this could assist researchers in the understanding of ACL injuries.

The isokinetic legpress exercise only measured forces in the sagittal plane. As a 180° change of direction occurs in all three planes of motion, future research should analyze the eccentric forces that occur in each respective plane of motion. Eccentric strength in the frontal plane of motion would be of particular interest, both from a performance- and injury prevention perspective.

Finally, this study only examined female soccer players. Future research should include male soccer players and analyze differences between groups (e.g. sex, limb dominance, strength

(35)

5. Conclusions

The main finding of this study was, contrary to previous investigations, that no significant relationship could be found between eccentric strength and change of direction performance, nor could it be found between eccentric strength and the different GRF parameters. No correlation was found between the GRF parameters and change of direction performance. However, a significant difference was found for eccentric peak force (p= 0,036) when examining the differences between the dominant and non-dominant limb. These results suggested that multi-joint eccentric strength has no correlation with change of direction performance. More research is needed to fully understand the role of eccentric strength for change of direction performance.

(36)

References

Andrews, J. R., McLeod, W. D., Ward, T., & Howard, K. (1977). The cutting mechanism. American Journal of Sports Medicine, 5(3), 111.

Arendt, E., & Dick, R. (1995). Knee injury patterns among men and women in collegiate basketball and soccer. NCAA data and review of literature. The American Journal of Sports Medicine, 23(6), 694–701.

Askling, C., Karlsson, J., & Thorstensson, A. (2003). Hamstring injury occurrence in elite soccer players after preseason strength training with eccentric overload. Scandinavian Journal of Medicine & Science in Sports, 13(4), 244–250.

Barnes, C., Archer, D. T., Hogg, B., Bush, M., & Bradley, P. S. (2014). The evolution of physical and technical performance parameters in the English Premier League. International Journal of Sports Medicine, 35(13), 1095–1100.

Bjordal, J. M., Arnły, F., Hannestad, B., & Strand, T. (1997). Epidemiology of anterior cruciate ligament injuries in soccer. The American Journal of Sports Medicine, 25(3), 341– 345.

Bloomfield, J., Polman, R., & O’Donoghue, P. (2007). Physical Demands of Different Positions in FA Premier League Soccer. Journal of Sports Science & Medicine, 6(1), 63–70.

Bobbert, M. F., Gerritsen, K. G., Litjens, M. C., & Van Soest, A. J. (1996). Why is countermovement jump height greater than squat jump height? Medicine and Science in Sports and Exercise, 28(11), 1402–1412.

Boden, B. P., Dean, G. S., Feagin, J. A., & Garrett, W. E. (2000). Mechanisms of anterior cruciate ligament injury. Orthopedics, 23(6), 573–578.

Bradley, P. S., & Noakes, T. D. (2013). Match running performance fluctuations in elite soccer: indicative of fatigue, pacing or situational influences? Journal of Sports Sciences, 31(15), 1627–1638.

(37)

Carling, C., Le Gall, F., & Dupont, G. (2012). Analysis of repeated high-intensity running performance in professional soccer. Journal of Sports Sciences, 30(4), 325–336.

Cronin, J. B., & Templeton, R. L. (2008). Timing light height affects sprint times. Journal of Strength and Conditioning Research, 22(1), 318–320.

Dai, B., Garrett, W. E., Gross, M. T., Padua, D. A., Queen, R. M., & Yu, B. (2015). The effects of 2 landing techniques on knee kinematics, kinetics, and performance during stop-jump and side-cutting tasks. The American Journal of Sports Medicine, 43(2), 466–474.

Dempsey, A. R., Lloyd, D. G., Elliott, B. C., Steele, J. R., & Munro, B. J. (2009). Changing sidestep cutting technique reduces knee valgus loading. The American Journal of Sports Medicine, 37(11), 2194–2200.

Dempsey, A. R., Lloyd, D. G., Elliott, B. C., Steele, J. R., Munro, B. J., & Russo, K. A. (2007). The effect of technique change on knee loads during sidestep cutting. Medicine and Science in Sports and Exercise, 39(10), 1765–1773.

Dirnberger, J., Huber, C., Hoop, D., Kösters, A., & Müller, E. (2013). Reproducibility of concentric and eccentric isokinetic multi-joint leg extension measurements using the IsoMed 2000-system. Isokinetics and Exercise Science, 21(3), 195–202.

Dirnberger, J., Kösters, A., & Müller, E. (2012). Concentric and eccentric isokinetic knee extension: A reproducibility study using the IsoMed 2000-dynamometer. Isokinetics and Exercise Science, 20(1), 31–35.

Dirnberger, J., Wiesinger, H.-P., & Müller, E. (2012). Reproducibility for isometric and isokinetic maximum knee flexion and extension measurements using the IsoMed 2000-dynamometer. Isokinet Exercise Science, 20(3), 149.

Dos’Santos, T., Thomas, C., Comfort, P., & Jones, P. A. (2018). The Effect of Angle and Velocity on Change of Direction Biomechanics: An Angle-Velocity Trade-Off. Sports Medicine, 48(10), 2235–2253.

The relationship between multi-joint eccentric strength and change of direction performance : A cross-sectional study on female soccer players